events that can be prevented at a population level through smoking cessation is substantial. 2 It is likely that selective attrition due to greater mortality in smokers explains, in part, the observed low prevalence of smoking, but with improved medical care, this survival difference may be decreasing; this sample experienced a change in demographics over time, with an increasing proportion of men in later surveys. Other factors that probably contributed to the low prevalence of smoking include implemented population-level interventions, such as legislation limiting smoking in public places and antismoking programs (e.g., ClearWay Minnesota).Limitations of the present study are the restriction to community-dwelling individuals, a possible underrepresentation of smokers due to volunteer bias, recall bias, decreasing response rate, modest sample size, and unclear generalizability to other geographic areas.In conclusion, between 1990 and 2009, the prevalence of current smoking in older adults remained consistently low as a result of quitting, antismoking programs, and probable effects of selective mortality in smokers. With an aging population, the absolute number of elderly smokers will probably increase despite its low prevalence.
ACKNOWLEDGMENTSConflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. *P < .05, † P < .01, ‡ P < .001; normocythemia versus polycythemia; chisquare test, Student t-test, logistic regression. § Adjusted for age, sex, obesity, hypoxia, race, and residence altitude using multiple logistic regression. SD = standard deviation; SpO 2 = oxygen saturation; BMI = body mass index; DM = diabetes mellitus; IFG = impaired fasting glucose.